def test_inv_index_query(load_inverted_index, wiki_docs):
words = ['after', 'were']
doc_ind = InvertedIndex.query(load_inverted_index, words)
assert {25, 290}.issubset(doc_ind)
words = ['neizvesnie', 'slova']
doc_ind = InvertedIndex.query(load_inverted_index, words)
assert len(doc_ind) == 0
words = ['after', 'were']
index_dict = build_inverted_index(wiki_docs)
inv_index = InvertedIndex(index_dict)
doc_ind = inv_index.query(words)
assert {25, 290}.issubset(doc_ind)
def test_query_2_intersect_words(words=['bow', 'tfidf']):
inv_idx = InvertedIndex(TEST_INDEX_TABLE)
doc_ids = inv_idx.query(words)
right_answer = {
4,
}
assert doc_ids == right_answer
def test_query2(mock_file):
""" Check if we can query some correct words with repeats. """
my_index = InvertedIndex()
my_index.build(mock_file)
assert my_index.query(['string', 'long']) == {1, 10, 100500}
def test_query_from_loaded2(mock_file):
""" Check if we can query some non-existed words. """
my_index = InvertedIndex()
my_index.load('test.index')
assert my_index.query(['me', 'test', 'non existed']) == set()
assert len(mock_file.mock_calls) > 1
def test_query_one_doc_in_index():
index = InvertedIndex()
index.inverted_index = defaultdict(set, {
'foo': {1, 2, 3},
'bar': {1},
'foobar': {1, 2}
})
assert index.query(['foo', 'bar'
]) == {1}, 'didnt find a doc, which present in index'
def test_query(mock_file):
""" Check if we can query some correct words. """
my_index = InvertedIndex()
my_index.build(mock_file)
assert my_index.query(['test']) == {
1,
}
def test_unicode_query_two_docs_in_index():
index = InvertedIndex()
index.inverted_index = defaultdict(set, {
'один': {1, 2, 3},
'bar': {1},
'два': {1, 2}
})
assert index.query(['один', 'два']) == {
1, 2
}, 'didnt find a two docs, which are present in index with unicode'
def test_query_from_loaded(mock_file):
""" Check if we can query after we load the index. """
my_index = InvertedIndex()
my_index.load('test.index')
assert my_index.query(['me', 'test']) == {
1,
}
assert len(mock_file.mock_calls) > 1
def test_unseen_word(word='fasttext'):
inv_idx = InvertedIndex(TEST_INDEX_TABLE)
doc_ids = inv_idx.query([word])
right_answer = None
assert doc_ids is right_answer
def test_query_2_words_without_shared_docs(words=['bow', 'cbow']):
inv_idx = InvertedIndex(TEST_INDEX_TABLE)
doc_ids = inv_idx.query(words)
right_answer = None
assert doc_ids is right_answer
def test_query_single_word(word='bow'):
inv_idx = InvertedIndex(TEST_INDEX_TABLE)
doc_ids = inv_idx.query([word])
right_answer = TEST_INDEX_TABLE[word]
assert doc_ids == right_answer
def test_query_not_in_index():
index = InvertedIndex()
index.inverted_index = defaultdict(set, {'foo': {1, 2, 3}, 'bar': {1}})
assert index.query(['foobar']) == set(), 'find a doc, that not in index'